Arch. lost. Razi, 1966, 18, 119 - 125

EXPERIMENTAL TRANSMISSION OF AFRICAN HORSE"SICKNESS BY MEANS OF MOSQUITOES ( * )

Y. Ozawa. G. Nakata, and F. Shad-del

SUMMARY

African horse-sickness was transmitted by means of the bites of stephensi aTld pipieTls which had engorged infected horse blood 15 to 22 days previously. Febrile response (40.4 to 40.5 C.) occurred approximately 16 days after the 2 horses used were bitten by infected mosquitoes. Death followed in 9 to 12 days. The cause of death was confirmed as African horse-sickness by postmortem findings and laboratory tests.

Introduction

It has long been known that African horse-sickness is not directly contagious. Many species of arthropods have been suspected as vectors of the because horse-sickness is usually most prevalent during warm wet seasons and can be pre­ vented by insect control.

As early as 1912, Schuberg and Kuhn 9 reported that they had transmitted horse-sickness mechan;cally by me ans of Stomoxys clllcitrans. Williams 11 reported Lypprosia minuta as a possible transmitter. Van Saceghem 10 suspected ticks and members of the order Diptera. Carpano 1 incriminated the dipterous genera. Anoph­ eles, , Phlebotomus, and S'imulillm as possible carriers of African horse-sick­ ness . The assertions of the latter 3 authors have been made without support­ ing experiment. Nieschulz et 111.,5 reported that Aedes mosquitoes were not significant vectors, as they could keep horse-sickness virus for only 1 week after experimental infec-

':' Reprint from the American Journal of Veterinary Research, Vol. 26, No. 112, 744-748

From the Near East Animal Health Institute, a cooperative United Nations Special Fund-FAO Project with the Ministry of Agriculture, Razi Serum and Institute, Teheran, Iran.

119 tion. Furthermore extensive investigations by Nieschultz et 111.6 on as potential victors of horse-sickness virus turned out to ,be negative, and they con­ cluded that mosquitoes were not vectors of horse-sickness. Du Toit 6 recovered African horse-sickness virus by inoculating horses with a suspension of Culicoidl's caught in the wild state. lt has not been determined how long ~ulicoides may carry the virus or whether Cl/timides may biologically transmit the 'disease trom horse to horse. Since this report, various species of Culimides have been widely believed to be the major vectors of African horse-sickness as sum­ marized in the recent report by Maurer and McCully.3 In the present experiment, ATlophel"s s(epheTlsi Liston and one of the most com­ mon mosquitoes, Culex pipiens, were used for transmission experiments and horse­ sickriess was biologically transmiUed to horses.

Materials and Methods

Mosquitoes.-Cull'x pipiens were caught in Teheran in April, 1963, and succes­ sive breedings were made at the inse:.:tarium of the Ne3r East Animal Health In­ stitute, Razi Institute. The 16th generation was used in the experiment. Anopheles s(ephl'Tlsi were also caught in Iran' The colony * had been main­ tained at the insectarium of the Institute of Parasitology and Malariology, Teheran University. Eight additional successive breedings were made before use in the ex­ periment. Horses.--Nonvaccinated horses, approximately 12 months old, were obtained from an isolated village in the mountainous area by the Caspian Sea. They were treated. with appropriate anthelmintics and kept under close observation for 1 week before' the experiment commenced. Virus.-Horse 20 was infected with African horse-sickness virus by intraven­ ously injecting the 5th mouse passage of Iranian strain 10/60. 9 The inoculum was 6,300,000 LD50 in suckling mice. Blood obtained from the horse 1 day before it died of horse-sickness (2 wk. after the injection of virus) was defibrinated and stored ,at 4 C. This blood was used as the material for infecting mosquitoes. Virus Isolation~Demonstration of virus from the blood of infected horses and from the tissues of dead horses was carried out by injecting the materials into the q~ains of 5-week-old mice. Defibrinated blood specimens were diluted with an

~, 'Pa.rtof the colony obtained through courtesy of Dr. Modlfl, Institute of Parasi­ tology, and Malariology, Teheran University, Teheran, Iran.

120 equal volume of saline solution. Tissues from infected horses :vere mi.xed with an equal volume of saline solution and ground in Ten Broeck tIssue gnnders. The tissue suspension was centrifuged at 2,000 r.p.m. for \0 minutes, and the super­ natant fluids were used as inoculum. Each specimen was injected into 10 mice, e3ch given 0.03 ml. of inoculum. The brains of infected mice in extremis were har­ vested and stored at -25 C. Neutralization Tests-Antibody levels in blood collected from infected and con­ trol horses were determined before and after infection. All the collected serums were heated at 56 C. for 30 minutes and the antibody titers measured by the standard virus dilution method described previously * * The same techniques were employed for identification of the virus isolated, using mice instead of tissue culture in tubes. Transmission Experiml'Tll.-The defibrinated blood of infected horse 20 was obtained on Feb. 25, 1964 (13 days postinoculation). Female mosquitoes of each of of the 2 species were allotted to 2 groups, A and B; the 4 groups of mosquitoes were kept in separate cages and fed with in­ fected blood on different days (Table 1). Before being fed the blood, al! the mos-

TABLE 1-Timetable of Successful TransmiSSion Experiments Incubation No. of pcriod of nl06quitocs Date ferl with virus in enJ;orged/ l\.lo~quito inCccted Dute Cod on mosquitocs* No. plaeed Horse ,g-roup hlood normal hOTSf"S (days) on horses No.

A 1HJ}J/I eles .del'hetlsi (A) :\[a1"('h 1-~ Mareh 18-19 15-18 14/30 TE-3 .11Jo}Jhele ..~ stephensi (13) Mareil 1-3

('1I'eI pi/,iens (A) Fen. 26-27 Mareh 18-19 15-22 2/19 TE-6 Cft/ex pipiens (B) lI!arch 2-:1

* ~Iinimum and maximum inlervnls bclwcen Ihe lst fecding with infected horse blood and tho 2nù fec(ling on the normal horses. quitoes were starved for more th an 8 hours by removing the water and 5% glucose wlution from the cages. Then, group A of each species was fed with undiluted in­ fected blood and group B with the blood diluted in an equal volume of physiologie saline solution. To feed the mosquitoes, cotton wool placed in small petri dishes was saturated by adding 12 ml. of either the diluted (B) or undiluted (A) infected horse blood. One petri dish was placed in each cage overnight for about 16 hours, after which the engorged mosquitoes were transferred to separate cages.

,•• , Hazrati, A., and Ozawa, Y.; Serological Studies of Afrlcan Horse-Slckness Virus-With Emphasis on Neutralization Test in Tissue Culture: Unpublishec;1 data, 1964,

121 These engorged mosquitoes were kept alive by feeding every day with fresh water and 5 % glucose solution. They were also fed calf serum twice a week. The temperature range of the insectarium was 23.0 to 26.5 c., and the range of relative humidity was 62 to 78%. Ali horses were maintained in insect-proof rooms throughout the experiment. At the end of an of 15 to 22 days, mosquitoes of group A and B of the same species were pooled and placed on a healthy horse by the fol­ lowing method: The mosquitoes were placed in a cage, 15.0 by 8.5 by 7.3 cm., made with a strong metal frame and fabric (organdy). The bottom of the cage was closely attached to the shaved skin of the horse's back. This small cage was covered with another stronger cage of metal with wire mesh. These cages were tightly fixed to horses with belts and springs for about 16 hours. During this period, the cages were covered with thick cio th to keep out the cold. Two additional horses were placed in the same stable during the experiment. One of them was actually exposed to infected mosquitoes, C. piPÙ'IIS, but none of the mosquitoes engorged the horse's blood. The other horse was kept without any treatment. Body temperatures of all horses were recorded twice daily throughout the ex­ periment

Results

African horse-sickness virus was in the defibrinated blood of horse 20 used to feed mosquitoes. The virus was identified as the Asia type virus by serum neutra­ lization tests. Virus was also found in the blood of hor,es TE-6, collected wh en they had revers above 40 C. and identified as the same Asia type of Afri:an horse-sickness virus by neutralization tests in mice. The course of the disease in infected horses was fol\owe:! by recording their body temperatures and any abnonnal signs. Horses 20, TE-6, and TE-3 had similar temDerature responses (Fig. 1), except that the temperature rise appeared severer in the ino:.:ulated hors 20 than in those horses infected by mosquito bites. AIso, the disease was much more protracted in the horses in­ fected by mosquito bites, and they lost a con,iderab'e amount of weight during the course of the di,ease. During the last 2 days be­ ,. fore death, these 3 horses could not 5J remain standing. The other 2 DA vs AFTER E XPOSURE horses, TE-I and TE-5, which were Fig. l-Postexposure temperatures

122 mosquito engorged blood, did not have signs of disease, and their body tempera­ tures remained between 37 and 39 C. throughout the experiment. Antibody responses were investigated in infe;;ted and control horses. Serums were coIle;;ted l'rom these ho;~es just before infection and 24 days after infection. The resuIts summarized (Table 2) indicate that no significant amount of antibody against Asia type horse-sickness virus had been produced in the infecte:! horses.

TABLE 2-Tests of Antibody Levels in the Serums of the Control ,and Infected Hon:es Before and 24 Days After Infection.

Horse No. Preinfection Postinfection NI TE-1 1063'" 106.3 0 TE-5 106.2 106.2 0 TE-3 1065 106.2 0.3 TE-6 106.2 105.5 0.7

,', Titers of virus-serum mixtures (TCID50jml. of the mixture).

The pcstmortem examinations of th:: 3 de'1d horses were made with Dr. Soh­ rab. who had con'Jucted numerous ne::ropsies on animaIs aFfected with African horse-sickness in Iran. The major gross lesions are summarized. Hors/' 20.-Tnere was subcutaneous yellow gelatinous infiltrate around the eyes and ears and in the neck, especially around the site of injection over the shoulder, and it was pronounced along the jugular furrow. The peri cardial sac containe.1 an excess of redd;sh yellow pericardial f1uid. The fat around the base 01' the heart was orange-yellow and gelatinous. and there were numerous petechiae on the heart surface. On section. hemorrhages were found in the myocardium. There were small focal are:lS of subendocarid'1l hemorrhage. The valve:; were thick and edematous. One of the lungs was slightly swollen and he'1v:er than the other. On cross section. the lung exuded a small amount 01' Frothy serous F1uid. The frot1y F1uid in the Iungs extended into the trachea. The gastric mucosa was hemorrhagic and acutely in­ flamed. The Fat covering the kidneys was edematous. On cross section. the corti­ comedullary area was extensive!y congested. Hors,' T.é··6.-Yellow, gelatinous, edematous infiltrate was mostly confined to the ventral muscles and hindlegs. The yellow ge!atinous infiltrate in the ventral muscles was most pronounced in an area be!ow the site where the mosquito cage was placed. Acute hemorrhage and necrotic inflammation were found in the in­ testine. The Fat covering the kidney œ'1rest the site 01' mosquito bites was gelatin­ ous. and the corticomedullary are'1 was ,everely congested. The changes in the heart and lungs were similar to those in horst:: 20, except the lungs were more

123 edematous and filled with frothy f1u:d, and sorne ye!lowish serous f1uid was in the bronchioles. Horse TE·~.- Yellow gelatinous infiltrate W:lS not p;trticubrly pron~Jlncej at the site of the mosquito bites but occurre1 mos!ly in the ve:t:ral muscles and legs. The eyelids also were edmatous and a slight amount of yellow ge!atinous infiltnte occurred around the eyes, in the region of t;1e trachea, und along the lumbar ver-

tebra. The lungs conta:ne:i less frothy fluid th an thc~e in the othe~ horses, but the other horses, but the interlobular septums stood out racher promine'ltly in sorne parts of the lungs. There was white froth in the cron~hial tubes. There were few spots of subendocardial hemorrhage in the heart. Other changes were generat:y the sa me as in the other 2 horses. These 3 horses are considered to have died of African horse-sickness.

Dis(u5:sion

Horses TE-3 and TE-6 die:! of African horse·s;ckness as determ' ncd by c!inical signs, postmortern changes, and virus recovery. That the disease was transmitted by the infe:ted mosquitoes is indicate:i by: (1) Colony-bred mosquitoes were me:! to avoid presen'.:e of viru, in the ime;;ts prior to the experiment, (2) the stabl:: in which the horses were iT''J.;ntained pr'J' vided strict insect-prcof isolation, (3) the characteristic pathologie ch:l!lges of horse­ sickness developed in the infected horses, a!11 (4·) t:e inlervals between the time of mm:quito bites and the 1st temperature reST)'J'lses are within a relso'lab'e period of incubation to indicate that the disease was transmitted by the 1T!0squito bites. The incubation periods of dis~aœ in the 2 hors es i:1fe:;'ed by rncquito bites were langer than those in artifically i!1fecte:! horse 20, possibly because of indiv!­ dual differences in susceptibiJity and a1so becaw:e of n d;fferc:Jce in the amount of virus introduced. The latter may also explain the dit-re~ences in the course of the disease. Ge'atinous infiltration of the fat was most prevalent in the hindquarters of infected horses and close to the sites of mm:quito bites. The experiment al

124 lor negative results pi"eviously obtained by other workers util;zing mosquitoes. 6 As in the czse with Japanese B eClcephalitis virus, 4 many transmission experiments were unsuccessful before positive results were obt3.ined with the same species of mosquito. Aceording to Nieschulz and Du Toit's experirrelt,6 the virus admin'stered to mosquitoes was detectab'e o:1ly for tI":e first 7 days. This is one reason why an interval of more than 15 days was cha~en befare permitting the infected mosqui­ toes to feed on normal horses in the p!e:-:elt exp~riment. Several factors remain to be c!arifjej as a ïesult of this transmission of Afriean horse-sickness by mosquitoes. It appe1rS t'lat: (1) The mosquitoes actej as biologie hosts rather than mechanica! tr3.nsT!1itter~: (2) virus ingested by the mosquitoes would have been either imctiv3.[ed or excreted together with the water anj glucase solution taken daily for ITIOïe th3.n 2 we~ks, thus e'imimting residu3.1 v'rus from their salivary glands; and (3) African t:cJïSe·sickness virus multiples in mosquitoes when they are kept under favorable cc 1d jt;ons.

REFERENCES

1. Carpano, M. : African Horse-Sickness as Observed Particularly in Egypt and Eritreo. Minist. of Agric. Tech. & Sci. Serv., Bull. 115, Cairo (1931) : 1-41. 2. Du Toit, R. M. : The Transmission of Blue·Tongue and Horse-sickness by ClllLcides. Onderstcpoort J. Veto Sci. & Anim. Indust., 19, (1944) : 7-16. 3. Maurer, F. D., and McCully, R. M. : African Horse-Sickness-With Emphasis on Pathology, Am. J. Veto Res., 24, (March, 1963) : 235-266. 4. Mitamura, A., and Kitaoka, M. : On the Epidemiology of Epidemie Encepha­ litis. Proc. XXIIth Jap. Med. Congress (1947) : 47-6. 5. Nieschulz, O., Bedford, G. A. H., and Du Toit, R. M. : Investigations into the Tra:lsmission of Horse-sickness at Onderstepoort du ring the Seaso:l 1931-1932 On .. derstepoort J. Veto Sci. & Anim. Indust., 3, (1934) : 275-344. 6. Nieschulz, O., and Du Toit, R. M. : Investigations i!lto the Transmission of Horse-sickness at Onderstepoort during the Seaso:l 1932-1933. Onderstepoort J. Veto Sci. & Anim. Indust., 8, (1937) : 213-268. 7. Ozawa, Y., and Hazrati, A.: Growth of African Horse-Sickness Virus in Monkey Kidney Cell Cultures. Am. J. Veto Res., 25, (March, 1964) : 505-511. 8. Rafyi, A. : La Pest Equine. Arch. Inst. Razi, 13, (1961) : 60-106. 9. Schuberg, A., and Kuhn, P. : Uber die Ubertrp,gung von Krankheiten durch cinheimische ::otechende Insekten. Arb. k. GS:ldhtsamt., 40, (1912) : 209-234. 10. Van Saceghem. R. : La Pest du Cheval ou Horse-Sickness au Congo Belge. Bull. Soc. path. exot., 11, (1918) : 423-432. 11. Williams, A. J. : Notes on an Out break of Horse-Sickness Connected with the Pre~ence of Lyperosia as Possible Transmitter. Veto J., 69, (1913) : 382-386.

125